Gabriel Wardi1, Morgan Carlile2, Andre Holder3, Supreeth Shashikumar4, Stephen R Hayden2, Shamim Nemati4. 1. Department of Emergency Medicine, University of California-San Diego, San Diego, CA; Division of Pulmonary, Critical Care, and Sleep Medicine, University of California-San Diego, San Diego, CA. Electronic address: gwardi@health.ucsd.edu. 2. Department of Emergency Medicine, University of California-San Diego, San Diego, CA. 3. Division of Pulmonary, Allergy, Critical Care and Sleep Medicine, Emory University, Atlanta, GA. 4. Department of Biomedical Informatics, University of California-San Diego, San Diego, CA.
Abstract
STUDY OBJECTIVE: Machine-learning algorithms allow improved prediction of sepsis syndromes in the emergency department (ED), using data from electronic medical records. Transfer learning, a new subfield of machine learning, allows generalizability of an algorithm across clinical sites. We aim to validate the Artificial Intelligence Sepsis Expert for the prediction of delayed septic shock in a cohort of patients treated in the ED and demonstrate the feasibility of transfer learning to improve external validity at a second site. METHODS: This was an observational cohort study using data from greater than 180,000 patients from 2 academic medical centers between 2014 and 2019, using multiple definitions of sepsis. The Artificial Intelligence Sepsis Expert algorithm was trained with 40 input variables at the development site to predict delayed septic shock (occurring greater than 4 hours after ED triage) at various prediction windows. We then validated the algorithm at a second site, using transfer learning to demonstrate generalizability of the algorithm. RESULTS: We identified 9,354 patients with severe sepsis, of whom 723 developed septic shock at least 4 hours after triage. The Artificial Intelligence Sepsis Expert algorithm demonstrated excellent area under the receiver operating characteristic curve (>0.8) at 8 and 12 hours for the prediction of delayed septic shock. Transfer learning significantly improved the test characteristics of the Artificial Intelligence Sepsis Expert algorithm and yielded comparable performance at the validation site. CONCLUSION: The Artificial Intelligence Sepsis Expert algorithm accurately predicted the development of delayed septic shock. The use of transfer learning allowed significantly improved external validity and generalizability at a second site. Future prospective studies are indicated to evaluate the clinical utility of this model.
STUDY OBJECTIVE: Machine-learning algorithms allow improved prediction of sepsis syndromes in the emergency department (ED), using data from electronic medical records. Transfer learning, a new subfield of machine learning, allows generalizability of an algorithm across clinical sites. We aim to validate the Artificial Intelligence Sepsis Expert for the prediction of delayed septic shock in a cohort of patients treated in the ED and demonstrate the feasibility of transfer learning to improve external validity at a second site. METHODS: This was an observational cohort study using data from greater than 180,000 patients from 2 academic medical centers between 2014 and 2019, using multiple definitions of sepsis. The Artificial Intelligence Sepsis Expert algorithm was trained with 40 input variables at the development site to predict delayed septic shock (occurring greater than 4 hours after ED triage) at various prediction windows. We then validated the algorithm at a second site, using transfer learning to demonstrate generalizability of the algorithm. RESULTS: We identified 9,354 patients with severe sepsis, of whom 723 developed septic shock at least 4 hours after triage. The Artificial Intelligence Sepsis Expert algorithm demonstrated excellent area under the receiver operating characteristic curve (>0.8) at 8 and 12 hours for the prediction of delayed septic shock. Transfer learning significantly improved the test characteristics of the Artificial Intelligence Sepsis Expert algorithm and yielded comparable performance at the validation site. CONCLUSION: The Artificial Intelligence Sepsis Expert algorithm accurately predicted the development of delayed septic shock. The use of transfer learning allowed significantly improved external validity and generalizability at a second site. Future prospective studies are indicated to evaluate the clinical utility of this model.
Authors: Mervyn Singer; Clifford S Deutschman; Christopher Warren Seymour; Manu Shankar-Hari; Djillali Annane; Michael Bauer; Rinaldo Bellomo; Gordon R Bernard; Jean-Daniel Chiche; Craig M Coopersmith; Richard S Hotchkiss; Mitchell M Levy; John C Marshall; Greg S Martin; Steven M Opal; Gordon D Rubenfeld; Tom van der Poll; Jean-Louis Vincent; Derek C Angus Journal: JAMA Date: 2016-02-23 Impact factor: 56.272
Authors: Seth W Glickman; Charles B Cairns; Ronny M Otero; Christopher W Woods; Ephraim L Tsalik; Raymond J Langley; Jennifer C van Velkinburgh; Lawrence P Park; Lawrence T Glickman; Vance G Fowler; Stephen F Kingsmore; Emanuel P Rivers Journal: Acad Emerg Med Date: 2010-04 Impact factor: 3.451
Authors: Gary B Smith; David R Prytherch; Paul Meredith; Paul E Schmidt; Peter I Featherstone Journal: Resuscitation Date: 2013-01-04 Impact factor: 5.262
Authors: Ryan C Arnold; Robert Sherwin; Nathan I Shapiro; Jennifer L O'Connor; Lindsey Glaspey; Sam Singh; Patrick Medado; Stephen Trzeciak; Alan E Jones Journal: Acad Emerg Med Date: 2013-05 Impact factor: 3.451
Authors: R Andrew Taylor; Joseph R Pare; Arjun K Venkatesh; Hani Mowafi; Edward R Melnick; William Fleischman; M Kennedy Hall Journal: Acad Emerg Med Date: 2016-02-13 Impact factor: 3.451
Authors: Andre L Holder; Namita Gupta; Elizabeth Lulaj; Miriam Furgiuele; Idaly Hidalgo; Michael P Jones; Tiphany Jolly; Paul Gennis; Adrienne Birnbaum Journal: Int J Emerg Med Date: 2016-02-24